Market Basket Analysis

Mahendra Gupta
Lecturer(CSE)
Govt. Polytechnic College Anuppur (M.P.)
 Contents
1. Introduction
2. Objective
3. Procedure( Apriori Algorithm)
4. Applications
5. Summarization
6. References

At NITTTR Bhopal, M.P. 2

 Introduction

 A large number of people use to visit

shopping malls and shopping centers[1].
 Massive amounts of data continuously
being collected and stored as
transactions
 Those collected data can be useful
from the business prospective such as
At NITTTR Bhopal, M.P. 3
 Objective

1. To find frequently purchased item

sets from large transactional
database[1].
2. Application.

At NITTTR Bhopal, M.P. 4

 Definition

 The process of discovering frequent

item sets in large transactional
database is called market basket
analysis[1].
 Frequent item set mining leads to the
discovery of associations and
correlations among items[1].
At NITTTR Bhopal, M.P. 5
The Apriori Algorithm[1]
(For finding frequent item sets)

Step:1 This step simply Table:1

scans all of the
transactions in order to
count the number of
occurrences of each
item, named it as C1.
Step:2 Apply minimum
support count, and
items set that satisfies
the condition named
as L1. (min support count
=2)

At NITTTR Bhopal, M.P. 6

Step:3 To discover the set of frequent 2-itemsets,
L2, the algorithm uses the join L1XL1 to generate a
candidate set of 2-itemsets, and count the
occurrences of each set.
Step:4 Apply min support count on C2, result will
be 2-frequent item set L2.

At NITTTR Bhopal, M.P. 7

Step:5 To generate 2-frequent item set, generate
3-item using L2XL2 and apply min support count
which will be 3-frequent item set.

Step:6 Continue until you get empty sets.

At NITTTR Bhopal, M.P. 8

 Application
1. Cross Selling of the Items[3][4]

At NITTTR Bhopal, M.P. 9

 Application
2. Proper Placement of Items[5].

3. Fraud detection.
4. Customer Behavior.
5. Affinity promotion[6].

At NITTTR Bhopal, M.P. 10

 Summary
 Market basket analysis.
 Discovering Frequent item sets.
 Application of frequent item sets.

At NITTTR Bhopal, M.P. 11

 References
[1] Jiwawi Han and Micheline Kamber, Concepts and Techniques of Data
Mining, 2nd ed., Jim Gray, Ed.: Morgon Kaufmann,Elsevier, 2012
[2] http://www.alamy.com/stock-photo/shopping-trolley.html
[3]http://www.amazon.in/
[4] http://search.shopclues.com/
[5] http://www.salemmarafi.com/code/market-basket-analysis-with-r/
[6]http://people.revoledu.com/kardi/tutorial/MarketBasket/Applications.htm
[7] http://youtube.com.

At NITTTR Bhopal, M.P. 12

At NITTTR Bhopal, M.P. 13